how i can do a speech diarization without know labels (unsupervised)?

hello everyone, i have a challenge that not do think how to solve the problem. I need to do a speech diarization in unknown duration time, and number of speakers, in European Portuguese. I have yet to try some possibilities (e.g., clustering with mfcc), but until now without success. Someone can help me solve this? Sorry for my bad English, and if this is a bad question to ask in this forum.

Felipe L. Teixeira

Answer 1

[Ishit]

Hi Felipe,

Please refer to the following links

1. Speaker Recognition and Diarization - MATLAB & Simulink - MathWorks India

2.unsupervised

3.GMM

4.Markov

[Felipe L. Teixeira]

@Ishit thanks. I have already studied some of these links, but I will read them again. thank you very much for your willingness to help